Coexisting depression among individuals abusing or dependent upon drugs has been associated with increases in risky drug-use-related and sexual activities. The specific pattern of this relationship of coexisting depression to risk has not been thoroughly explored, particularly in a longitudinal framework. This study seeks to illuminate the relationship across time between high-risk behavior and depressive symptomatology among a sample of injection drug users not currently in treatment. Through analyses of a unique longitudinal database (described below), we will address three specific aims, illuminating the relationship between coexisting drug use, depressive symptomatology, and risk behavior. First, through a factor analysis of the BDI-2 and CES-D, we will determine the underlying factor structure of depressive symptomatology among this sample of drug users. Second, we will assess the relationship between the identified depressive symptomatology factors and high-risk sexual activity, substance use patterns, risky drug-use-related behaviors, and HAV, HBV, and HCV serostatus. Third, we will determine whether across time, changes in depressive symptomatology lead to changes in and/or can successfully predict sexual risk behaviors, drug-use-related risk behaviors, substance use patterns, and HAV, HBV, and HCV serostatus. We will base our analyses on a longitudinal database established through a NIDA funded project implemented in Anchorage, Alaska from 1997 to 2000. This randomized control clinical trial sought to determine whether participation in a needle exchange program would result in increased injection behavior and its sequellae. To accomplish this, 600 injection drug users (IDUs) were randomly assigned to one of two conditions (access to a needle exchange program or pharmacy sales). Of these participants, 563 completed the Beck Depression Inventory-2 (BDI-2) and the Center for Epidemiological Studies - Depression scale (CES-D) at three points in time: baseline, 6-month, and 12-month follow-up. Extensive risk assessments, using the Risk Behavior Assessment (RBA) at baseline and the Risk Behavior Follow-Up Assessment (RBFA) at 6- and 12-month follow-up, were collected. At all three points of contact, participants were tested for HAV, HBV, and HCV serostatus. The resulting longitudinal database is of sufficient size to support the complex analyses proposed to answer the questions posed in our three specific aims. [unreadable] [unreadable]